苹果研究:用户不愿被AI代替决策

鞭牛士
Feb 13

AIPress.com.cn报道

在AI代理能力不断增强的背景下,Apple研究团队将焦点转向一个更具现实意义的问题:当AI可以替用户“操作电脑”时,用户到底愿意放权到什么程度?

该研究首先分析了9款现有代理系统,包括Claude Computer Use Tool、OpenAI Operator、Project Mariner等,构建出覆盖“输入方式”“行为可解释性”“用户控制”“心智模型”四大维度的体验框架。

研究指出,当前AI代理设计普遍强调自动化与执行能力,却未充分回答一个核心问题——用户如何实时理解代理的决策逻辑。

在随后的用户实验中,20名参与者通过聊天界面与“代理”完成度假租赁与电商购物任务。实际上,所谓代理由研究人员人工模拟操作。

实验结果呈现出清晰趋势:

用户希望看到代理在做什么,但不希望被迫逐步监管

在探索型任务中容忍更高自主度

在执行型任务中更强调精准与一致

在涉及金钱或账户变更时,强烈要求确认机制

最具代表性的发现是,一旦代理在模糊选项下自行做出未说明的选择,参与者普遍要求系统暂停并解释理由。若代理悄然偏离原计划,信任度迅速下降。

研究认为,AI代理的关键挑战并非能力不足,而是“黑箱决策”带来的心理落差。当系统代替用户点击、选择、提交信息时,任何未被解释的假设都可能被视为风险。

这项研究为AI代理的产品化路径提供了更明确方向:真正可落地的代理系统,需要在自动化与可控性之间找到动态平衡,而非单纯追求更强的自主执行能力。(AI普瑞斯编译)

海量资讯、精准解读,尽在新浪财经APP

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10